PLSM3 status: ETA = When it's done.

tuan kuranes

07-01-2008 09:26:12

PLSM3 news:

Still working on it, but honestly very slowly, plus I tried many experimentations things, and therefore tested and threw many implementations and time...

I may perhaps tend to release a less featured PLSM3 or 2.5 , without the terrain renderer plugin diversity I'm after in PLSM3. (which is the biggest problem I have here.)

Problem is I'm under load of work lately, so working on it was very hard...
So still hard to say when, but as real life work is planned (very hard), not to be released before a few month or so.

I guess I should have have announced PLSM3 once ready instead of blocking/deceiving expectations.

Sorry again.

Still PLSM2 is a viable solution, even if needing some documentations and perhaps some patches.

As I'm very busy, new maintainers/patchers, forum moderators, or documenters are and would be very welcome.

beaugard

09-01-2008 20:56:57

Did you work on the CHC implementation (Occlusion octree scene manager...)? I have tested the existing implementation by pulling out the relevant files from plsm2, and while it does work (CHC works, pure frustrum culling doesn't) I think it could be done ... better :). The obvious thing is in the todo of plsm3 - to make it a project of its own. Another thing you have mentioned in the forums is to make use of a custom renderqueue...? There are more things that I don't remember.

In short, if you have worked on the Occlusion Octree SM, it would be of great value if you released it (not only to me, I think). If you haven't I'll start working on the sources from plsm2 (I understand if you haven't - maybe not the most interesting thing to spend your time on).


EDIT:
oh, and take the time you need, release what you feel you can release, etc. plsm2 is an amazing piece of work and developing anything on the side of work&family is always going to be slow. So don't feel bad about it...

tuan kuranes

10-01-2008 10:02:27

@beaugard: FYI: CHC released here.. As you'll see still need work into it, but I'm focusing on PLSM3.

beaugard

10-01-2008 13:18:58

Wonderful! Thank you very much.

Arroy

07-03-2008 08:03:16

how long for release? :)

btmorex

30-03-2008 20:47:47

I'd be interested in a PLSM 2.5. I wouldn't worry too much about releasing something that's not feature complete or undocumented.

Highway

26-05-2008 11:21:42

What is the development status on PSLM3/2.5 ? Do you still find time to work on it?

MOROHEUZ

29-05-2008 21:39:02

Good question ... ;-)
Unfortunately, I can no longer compile .. (Vc9 + + with Vista)

I think the day will come Shoggoth, perhaps there the whole nen upswing again ...

Unfortunately, this is nothing quite so easy ....

But perhaps SPT2 the day comes out ... that's enough to plms2 ran very near ... and if there is a terrain editor would be ... OHhhhhhoooo .. then its the best tool ive ever seen ;-)

Seregvan

29-10-2008 09:28:52

Hi all, any updates in PLSM3?

compvis

12-01-2009 09:46:04

Hello,

When will PLSM3 be released ? I can't wait longer !!! :mrgreen:

Ralith

12-02-2009 09:41:35

Any news? PLSM2 seems to have fallen out of sync with Ogre SVN.

DanielSefton

12-02-2009 12:31:11

Just to say, PLSM2 still works absolutely fine with the latest Ogre. If you're having problems, it's not because they are incompatible.

I'm sure Tuan will post an update when he feels it's ready -- he's a busy guy. :wink:

ERSmith

12-02-2009 16:24:12

Oh really? I've been wrestling with this all week and since the compile comes out full of "bla bla is not a member of Ogre::" it seems to indicate to me that plsm2 is trying to use nonexistent members.

tuan kuranes

12-02-2009 17:33:10

plsm3 is really near now, but still under heavy work.

Meanwhile make sure to use Ogre SDK or OGRE SVN tag 1.6 (I'm not supporting ogre SVN Trunk) AND plsm2 SVN

DanielSefton

12-02-2009 17:50:07

Oh really? I've been wrestling with this all week and since the compile comes out full of "bla bla is not a member of Ogre::" it seems to indicate to me that plsm2 is trying to use nonexistent members.
Yes... I got it working with the latest Ogre (of which I mean the 1.6.1 branch) straight away. Something must not be right in your setup. Btw, start a fresh project, the supplied project files are a bit dodgy. But post your problem; it's no good stewing over something for a week without help. :P

Ralith

13-02-2009 00:40:01

Moving back to Ogre 1.6, then, since I don't really have any reason to be using SVN.

By the way, anybody know if it would be practical to implement infinite terrain with PLSM2? Or 3, even, if it's that near? (i.e. load tiles from a procedural source of some kind rather than from disk)

stealth977

13-02-2009 08:52:24

Just out of curiosity, why do you need infinite terrain? Even, runtime procedurally created terrain?
- How will you make the coverage maps?
- How will you place entities on it?

Even for a Flight Simulator, it would be a dull terrain....

Stealth,

Wretched_Wyx

13-02-2009 17:16:19

Paging terrain is useful for lots of stuff, but mainly MMOs / large world oriented games and apps.

For example, in the game engine I'm working on, my goal is to implement a terrain system similar to how Oblivion, Fallout 3, the older TES games, etc do it. You have your main land masses, but modders can add onto the "world" easily, adding new land masses outside of the existing one. Thus, extending the world. So while initially a "world" might be a small island, it's easily made larger by adding more land mass. Having a paging terrain system makes something like this easy.

I have to agree on your point though, I see where you are coming from. I've read many articles on this topic, and I definitely agree that a smaller world packed with more detail, beats a large one devoid of life. Hence the reason I'll be taking the approach of "start small and expand outward".

[edit]
A good example of why such terrain systems are very useful, and can be put to good use: http://www.infinity-universe.com/Infinity/
[/edit]

Ralith

14-02-2009 01:37:33

Just out of curiosity, why do you need infinite terrain? Even, runtime procedurally created terrain?
- How will you make the coverage maps?
- How will you place entities on it?

Even for a Flight Simulator, it would be a dull terrain....

Stealth,

I plan to use boredom as an alternative to an invisible wall-of-death at the edge of the world. Runtime procedurally created terrain because a simple noise-based heightmap is nicer (and easier to improve later) than something tiled from disk.

Zero23

23-03-2009 00:59:49

What is the status?
Why do you not upload PLSM3 as branch in SVN, so that other people can help you?

superflanker

02-04-2009 17:59:28

Refute this, but all indications seem to indicate PLSM is dead.

See my other thread, the demo is bugged and I traced it, the problem is occuring somewhere in the plugin. Maybe it worked at one point in time, but
definately its broken.

Straight from the wiki
Latest win32 Binary Demo and sample Maps is here (03/07/2005) :shock:

I think a freshman in 2005 is due to graduate in a few months.

JeDi

03-04-2009 14:27:44

With all respect, but you seem to know tuan very well for someone who joined the forum only three days ago...

The author of this plugin is an active, well-known and much respected member of this community, so I am sure PLSM3 is going to rock very hard. As he stated, he already has a day-time job, and combining such a complex implementation with work and family is not trivial. I'm sure if someone would pay him enough to work on this fulltime, results would be ready a lot sooner :wink:

superflanker

03-04-2009 16:04:25

With all respect, but you seem to know tuan very well for someone who joined the forum only three days ago...

The author of this plugin is an active, well-known and much respected member of this community, so I am sure PLSM3 is going to rock very hard. As he stated, he already has a day-time job, and combining such a complex implementation with work and family is not trivial. I'm sure if someone would pay him enough to work on this fulltime, results would be ready a lot sooner :wink:


Well I don't need to know Tuan to see that the demo doesn't work. It is true 1 person cannont maintain software like this, but one of the principles of open-source is that there are many developers contributing to a project. So your point about being busy with work/family is moot. If Tuan cannot maintain semi-regular releases, he should use his leadership and find other developers to help out. I may sound a bit presumptuous, but the fact is a broken demo that is showcasing the tech is unacceptable for any self-respecting project. I've downloaded 5 other plugins, I'm only having problems with this one. I've noticed things related to this mod take on the orders of MONTHS and YEARS.

stealth977

05-04-2009 14:31:15


Well I don't need to know Tuan to see that the demo doesn't work. It is true 1 person cannont maintain software like this, but one of the principles of open-source is that there are many developers contributing to a project. So your point about being busy with work/family is moot. If Tuan cannot maintain semi-regular releases, he should use his leadership and find other developers to help out. I may sound a bit presumptuous, but the fact is a broken demo that is showcasing the tech is unacceptable for any self-respecting project. I've downloaded 5 other plugins, I'm only having problems with this one. I've noticed things related to this mod take on the orders of MONTHS and YEARS.


Some things to consider:
1 - Yes the demo is buggy, probably it was not synced with the plugin so some features of demo doesnt work...
2 - But the plguin works, you just need to spend some time on it
3 - PLSM3 may not be realized at all and thats not only because tuan is very busy but also because sinbad is working on something which will make nearly all terrain renderers obsolete (Paged World/Terrain - will be a core feature - check Development Thread) and the best part is he already started working on it and probably will finish it in very short time since it is also a part of his contract work...
4 - Still There is only one point superflanker's idea makes sense and that is: I saw many and I express MANY good projects which went dead in the community. All had one thing in common : They were all ONE MAN PROJECTS. One man projects are doomed to fail not only because they will lack to satisfy ever growing needs but also will vanish once its maintainer has no time to spend or has lost the will to go on with it... (Yes I know there are amazing One Man Projects in the community and I really appreciate their authors for their dedication)

ismail,

jacmoe

05-04-2009 15:10:49

Well I don't need to know Tuan to see that the demo doesn't work. It is true 1 person cannont maintain software like this, but one of the principles of open-source is that there are many developers contributing to a project. So your point about being busy with work/family is moot. If Tuan cannot maintain semi-regular releases, he should use his leadership and find other developers to help out. I may sound a bit presumptuous, but the fact is a broken demo that is showcasing the tech is unacceptable for any self-respecting project. I've downloaded 5 other plugins, I'm only having problems with this one. I've noticed things related to this mod take on the orders of MONTHS and YEARS.
Tuan did actually ask for other devs, several times.
None stepped forward.
Tuan's been frank about the current status of PLSM2 - you can read about it in the other stickies if you care.

DanielSefton

05-04-2009 22:11:25

/* when invited to a free lunch, you should not complain about the quality of the caviar */
:wink:

I'm just grateful a paging terrain plugin exists. No one is being paid for these projects. Not forgetting the fact that writing an advanced paging terrain system is an extremely difficult piece of software engineering. Besides; under the rubble, PLSM2 works perfectly, and is in fact one of if not the best Ogre plugins.

stealth977

06-04-2009 11:35:40


I'm just grateful a paging terrain plugin exists. No one is being paid for these projects. Not forgetting the fact that writing an advanced paging terrain system is an extremely difficult piece of software engineering. Besides; under the rubble, PLSM2 works perfectly, and is in fact one of if not the best Ogre plugins.


I second that :)

kungfoomasta

07-04-2009 00:10:22

Maybe PLSM2 underwent a lot of modifications after I gave it a go, but I find it hard to label PLSM2 as "works perfectly". I'm not trying to be insulting, just want clarification, as others who haven't used PLSM2 are reading this to find out what they're getting into. In my experiences, PLSM2 worked, but required a bit of tweaking and digging through the code, because there isn't a lot of public knowledge on the inner workings, and there is little activity on the project, in terms of development, bug fixes, etc. Again, I used PLSM2 a ways back, so things may have changed.. and just to be clear, I think its really great Tuan was generous enough to give PLSM to the community, I'm sure it has helped a lot of people.

jacmoe

07-04-2009 01:01:49

It is true 1 person cannont maintain software like this, but one of the principles of open-source is that there are many developers contributing to a project.
It is.
In an ideal world. :wink:

Unfortunately, noone has been able to step up to the challenge. And I blame noone, really.
That's just how it works in real life.

The other addons you mention have been blessed with contributers, and even additional developers joining.
Which, of course, is a good thing. :)

compvis

30-05-2009 00:55:19

The PLSM3 have died ? Long time no see Sir Tuan on Ogre3d community....

tuan kuranes

31-05-2010 12:30:52

It's not dead.
On hold of completion would be a better term.

mary

08-08-2016 11:36:59

Did you work on the CHC implementation (Occlusion octree scene manager...)? I have tested the existing implementation by pulling out the relevant files from plsm2, and while it does work (CHC works, pure frustrum culling doesn't) I think it could be done ... better :). The obvious thing is in the todo of plsm3 - to make it a project of its own. Another thing you have mentioned in the forums is to make use of a custom renderqueue...? There are more things that I don't remember.